Anyone Can Buy a Click: How to Read Your Own Ad Report

A real agency report, taken apart line by line – what the brand campaign hides, and the one question to ask whoever runs your ads.

“Can you look at this and tell me if it’s any good?”

A client forwarded me a marketing proposal a while back and asked me to poke holes in it.

This happens more than you’d think. People don’t usually call their web guy to vet a five-figure ad contract from another company… but they call me. (Turns out trust is a funny thing – once someone hands you the keys to their whole online presence, they start asking what you make of everything else too.)

So I read it. And I want to walk you through what I found, because it’s the same story almost every time.

What follows is a real report from a real agency, with the names taken off. I’ll show you the headline numbers, then show you the arithmetic that takes them apart – the same arithmetic you can run on your own report in about five minutes. At the end there’s one question to ask whoever runs your ads.

Top performer: 550 conversions at $2.11 each. It’s the brand name. The agency boxed it out themselves – hiding their best-looking number, because of where it came from.

Now look closer.

Look at the headline numbers first – 135,000 impressions, 14,000 clicks, a 10.55% click-through rate. Looks great on a Monday morning. The client sees big numbers and a green arrow and feels like the money’s working.

The strongest line in the whole account is a campaign built on the client’s own brand name. A 20.87% click-through rate. Twenty percent. That’s not marketing – that’s people typing the company’s name into Google because they already knew the company… and the agency bought an ad in front of them and took the credit. They covered it with a white block before sending the report.

Any brand can get that click-through rate. Any of them. You don’t need a contract in another state to capture people who were already searching for you by name – they were going to find you anyway. The agency just stepped in front of the door they were already walking through and stamped its name on the welcome mat.

Not all brand bidding is bad. Used to pad the numbers, it’s deceptive. Brand campaigns have their place – defending against competitors bidding on your name, controlling which page the click lands on. What they shouldn’t be is the centerpiece of an ROI story when the customer was already searching for you by name.

Do the arithmetic yourself. It takes five minutes.

Here’s the part nobody hands you, so I’ll hand it to you. You don’t need a marketing degree to check this. You need a calculator and the keyword-level report – not the campaign summary, the keyword rows. Find the line that is your company name. Subtract it from the total. Whatever’s left is the work you’re paying for.

Run it on the report above:

• Reported totals – 135,297 impressions, 14,271 clicks, 1,567 conversions, $20,053 spent.
• The brand row – 21,375 impressions, 4,460 clicks, 550 conversions, roughly $1,160 spent at 26 cents a click.
• Everything else – 113,922 impressions, 9,811 clicks, 1,017 conversions, roughly $18,893 spent.
• The real numbers – an 8.61% click-through rate, $1.93 a click, and $18.57 per conversion.

The report says $12.79 per conversion. Take the brand name out and it’s $18.57 – forty-five percent higher than the number on the cover page. The reported cost per click goes from $1.41 to $1.93. The reported click-through rate drops from 10.55% to 8.61%.

Nothing was falsified. Every number in that report is correct. The brand campaign is 5.8% of the spend and 35% of the conversions, and averaging it in with everything else drags the whole account’s numbers down to something that looks lean. That’s not a rounding error. That’s the difference between a campaign that’s working and one that needs a conversation.

And this is the version I can check. Most of the time nobody runs the subtraction, because the keyword-level report never gets sent – the client gets a slide with four big numbers on it and a green arrow.

Ask for the keyword rows. Every month.

A floor is not a guarantee.

There’s a line I keep coming back to. They guaranteed – in writing – a “minimum .1% CTR.”

Here’s the writing. A guaranteed floor, dressed up as a promise.

That’s one click per thousand views. It’s the floor. You basically cannot do worse than that. Guaranteeing a 0.1% click-through rate is like a restaurant guaranteeing the food won’t be frozen… technically a promise, just not one worth paying for.

Any guarantee written against clicks or impressions is a guarantee about activity, not outcomes. Impressions are inventory. Clicks are traffic. Neither one is a customer. The only number worth putting in a contract is cost per qualified lead, and almost nobody will sign that, because it depends on the one thing they don’t control…

…which brings us to the part nobody in these meetings ever brings up.

Almost none of these companies touch the website.

I’d know – I host and manage the sites for most of these clients, so I see exactly what changes and when. More often than not, the answer is nothing. No new headline to match the campaign. No page built for the people the ads are sending over. No change to the words, no change to the layout, nothing that turns a visitor into a phone call. They buy the click, drop the person on a page nobody looked at, and that’s the job.

And there’s a reason they leave it alone. Changing that page costs time and money – they don’t want to incur those costs themselves, they’d rather you spend it with them than with another vendor. Because a dollar in someone else’s pocket is a dollar that could have been in theirs.

Insider perspective: If I had a nickel for every time an ad agency expected a client’s web developer to build custom landing pages for free just so they could maximize their ad spend, I’d have a remarkably heavy bag of nickels.

So the one page that decides whether any of this even works is the one thing they will not touch. And touching it slows them down – it’s one more step between the signed contract and the cash register. Easier to start the meter today and leave the page right where they found it.

“This page is broken, it won’t convert.”

When they do say something about the site, it’s usually some version of that. Half right – it won’t. But not because it’s broken. Because it was never built to convert. An informational page and a landing page do two different jobs, and pointing an ad at the wrong one doesn’t make the page broken… it makes it the wrong page.

Here’s the difference, since nobody ever explains it.

An informational page is built for a general visitor who found you however they found you. It tells them who you are, gives them the tour, and links out to everything else on the site so they can wander. Its job is breadth.

A landing page is built for one specific person arriving from one specific ad. It repeats the promise the ad made, in the same words, above the fold. It has one action on it and most of the exits removed – no menu full of distractions, no “learn more about our history.” Its job is depth on a single question: do I call these people or not?

That distinction is where the money is. WordStream’s account-level data puts the median Google Ads landing page at a 2.35% conversion rate, while the top 10% convert at 11.45% or better – roughly five times the middle of the pack, on the same clicks, from the same ads. The gap isn’t the ad. The gap is the page.

And the trend is moving in the wrong direction for anyone who ignores it. Across 2026 benchmarks, click-through rates went up while conversion rates fell in 13 of 14 tracked industries. More people are clicking. Fewer of them are converting. That combination points at the page, not the ad – which is where I keep finding it. (I’ve written about the small points of friction that cost you conversions, and none of them are ad settings.)

The page isn’t broken. It’s the wrong page for the job.

Then there’s the version that should worry you the most.

Sometimes they don’t send the traffic to your website at all. They build their own page – on their platform, their CMS, their walled garden – and funnel every lead into their own dashboard, where you don’t control the page, the analytics, or what happens to that data after submission. Looks slick. Works fine. Right up until the day you cancel.

Then it all goes dark. The page vanishes. The leads vanish. The traffic was never yours, the remarketing list was never yours, the clicks were never yours. You rented an audience and the landlord kept the keys. You walk away with nothing to show for the spend except a stack of pretty PDFs… and you can’t even run an ad to the people who already raised their hand, because you never owned the click to begin with.

Six questions, before you sign anything:

• Whose Google Ads account is this? If it’s theirs, the entire spend history and every audience you built leaves with them.
• Where do the landing pages live? On my domain, or on their platform? If it’s not my domain, it’s not my page.
• How do leads reach me, and can I export them? A dashboard login is not ownership. Email and a CSV export are.
• Will I get keyword-level reporting? Campaign summaries hide brand traffic. Keyword rows don’t.
• Is brand traffic reported separately? If it’s blended into the totals, every number you’re shown is flattered.
• Who is changing my website to match the campaign? If the answer is nobody, you’ve bought traffic with nowhere to go.

None of those are hostile questions. Any operator running things properly will answer all six without blinking. The ones who get uncomfortable are telling you something.

Cheap brand clicks drag the average down. $1.41 looks lean until you see the 26-cent keyword doing all the work.

So what are these clients paying for?

It all comes back to the same thing. Anyone can buy a click – that’s the easy part. You hand a platform a credit card and it sends you traffic. The hard part, the part that decides whether any of it becomes a customer, is everything that happens after the click lands. And that’s the exact part every one of these setups leaves alone.

Clicks, pretty PDFs and flashy dashboards. Real work, sure – just the narrow part, and never the part that turns into money.

The guy who spent fifteen grand over three months and told me he “got nothing” wasn’t exaggerating. He got traffic. He just never got anything waiting to catch it. …and the website looked exactly the same on day ninety as it did on day one.

I’m not writing this to dunk on anybody. The platforms do real things, there are good operators out there, and plenty of times this works exactly like it should – Google Ads is the modern Yellow Pages and it earns its keep. Some clients spend $3,000 and pull $40,000 in new business out of it. At that point the brand-bidding line item is noise. That’s not a problem to fix, it’s a machine to scale… if it ain’t broke, leave it running.

In fairness to brand bidding: Google’s own research across hundreds of accounts found the large majority of mobile search ad clicks are incremental, and there are sound reasons to defend your name in the auction when competitors are bidding on it. The point isn’t that brand campaigns are fraud. The point is that they belong on their own line, not blended into the headline. Procter and Gamble cut brand search spend hard in a well-documented test and saw no change in business outcomes – and P&G can afford to find that out. Most small businesses can’t.

But if it is broke – if the phone’s quiet and the spend keeps climbing – you should know why, and you should know where your dollars are going. The gap is simple once you see it, and almost nobody points it out from the inside.

What “touching the website” actually means.

Since I’ve spent this whole article saying nobody does it, here’s what it is. None of this is exotic. It’s a few hours of work that turns bought traffic into a conversation.

Match the page to the ad. If the ad says emergency service, the page says emergency service, in the same words, at the top. Message match is the cheapest conversion gain there is and it costs nothing but attention.

Build a page per service, not a page for everything. A visitor who searched for one thing should land on that thing. A well-structured site already has these pages – they just need the ad copy pointed at them and the exits tightened.

Make the action obvious. Phone number visible without scrolling, tap-to-call on mobile, a short form, and nothing else competing for the eye. If someone has to hunt for how to contact you, they won’t.

Make it load. A slow page loses people before the copy has a chance. You paid for that click either way.

Then measure the page, not the campaign. Conversion tracking wired to real form submissions and real calls, so you’re grading the site instead of grading the ad platform’s homework.

That’s the work. It’s the reason my monthly marketing plan includes the site itself instead of stopping at the ad account – the page and the campaign are the same job, and splitting them across two vendors is how the gap in this article gets created in the first place. If the site is the part that’s broken, that’s fixable too, usually faster and cheaper than anyone expects.

And if the campaign’s fine and the site is the problem, you’d want to know that before spending another month on clicks. That’s the same conversation as “we don’t really use the website, it doesn’t do anything for us” – different door, same room.

One question.

So if you’re paying someone to run your ads, ask them this: when’s the last time you changed something on my website to convert the traffic you’re sending me?

If the answer is never – and it’s almost always never – you already know. You’re not paying for customers. You’re paying for clicks.

And anyone can buy a click.

Need Help?

Send me your last marketing report and I’ll tell you what you’re really getting. No pitch, no obligation – same read I gave the client in this article. If the campaign’s working, I’ll tell you that too, and you can go back to your day with one less thing to wonder about. Reach out anytime.

“You’re not paying for customers. You’re paying for clicks.”